1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
|
\name{SH.test}
\alias{SH.test}
\title{Shimodaira-Hasegawa Test}
\usage{
SH.test(..., B = 10000, data=NULL)
}
\arguments{
\item{...}{either a series of objects of class \code{"pml"}
separated by commas, a list containing such objects
or an object of class \code{"pmlPart"}.}
\item{B}{the number of bootstrap replicates.}
\item{data}{an object of class \code{"phyDat"}.}
}
\description{
This function computes the Shimodaira--Hasegawa test for a set of
trees.
}
\value{
a numeric vector with the P-value associated with each tree given in
\code{...}.
}
\references{
Shimodaira, H. and Hasegawa, M. (1999) Multiple comparisons of
log-likelihoods with applications to phylogenetic
inference. \emph{Molecular Biology and Evolution}, \bold{16},
1114--1116.
}
\author{Klaus Schliep \email{klaus.schliep@gmail.com}}
\seealso{
\code{\link{pml}}, \code{\link{pmlPart}}, \code{\link{pmlCluster}}, \code{\link{SOWH.test}}
}
\examples{
data(Laurasiatherian)
dm <- dist.logDet(Laurasiatherian)
tree1 <- NJ(dm)
tree2 <- unroot(upgma(dm))
fit1 <- pml(tree1, Laurasiatherian)
fit2 <- pml(tree2, Laurasiatherian)
fit1 <- optim.pml(fit1) # optimize edge weights
fit2 <- optim.pml(fit2)
SH.test(fit1, fit2, B=500)
# in real analysis use larger B, e.g. 10000
\dontrun{
example(pmlPart)
SH.test(sp, B=1000)
}
}
\keyword{models}
|
